Topic: MATRIX WITH THE UA-5

[jctibbitts]

Wow,

Sorry i didn't get back to this thread quickly...was in boston for some band !

anyway, thanks all for the advice.  my only problem with recording onto two devices and matrixing after the fact is i hate recording with the MD, can't afford a DAT or another JB3.

what if i had the NT4 as a stage mic per say, and also did the board feed into the UA5?  would i get delay there?  i guess i just have to test it out, but i wanted to see if anyone has done it.

PS.  how do you do a matrix after the fact?  with soundforge/other program?  

[Simp-Dawg]

Wow,

Sorry i didn't get back to this thread quickly...was in boston for some band !

anyway, thanks all for the advice.  my only problem with recording onto two devices and matrixing after the fact is i hate recording with the MD, can't afford a DAT or another JB3.

what if i had the NT4 as a stage mic per say, and also did the board feed into the UA5?  would i get delay there?  i guess i just have to test it out, but i wanted to see if anyone has done it.

PS.  how do you do a matrix after the fact?  with soundforge/other program?  

if you had the nt4 on stage you would not have to worry about delay issues with the board feed.  you'd just have to worry about all that cable getting from the sound board to your ua-5 (or mics to ua-5 depending where you put the rig!)

and post matrix can be done in wavelab montage or cool edit pro i think, probably others too.

so did you tape phish last night??  if so how did it come out?

[jctibbitts]

Didn't tape phish.  i just tape for my band.  i am still learning and would feel like a retard taping phish.  plus, i would be worrying about my gear the whole time rather than air guitaring and getting out of control.

i recommend everyone check out the rnr > groove>tweeprise>frank>kunk>frank

sick

[jctibbitts]

I guess my last question would be:

would matrixing before the fact (source i mentioned earlier) fix the stage volume problem i have with some boards?

i never put my mics on the stage because i would get no vocals.  then i tried just straight soundboard and at most venues i get too much vocals and no bass (stage volumes differ).  i just want a perfect clear mixed recording and want to know if it's possible with what i have:

nt4
jb3
ua5 wmod (will be buying very soon)

[Simp-Dawg]

think about what you just said...onstage lacks vocals...sbd has too much vocals...mix them together and voila!  you could do a post matrix but if you don't have to worry about time delay (by being on stage) and you have enough cabling to do it, i say go for the matrix with the ua5.
ka-bling.

[Brian]

shlading...what simp-dawg said. +t running the matrix on the fly is the best bet because once you get the levels down, you will be able to make some great tapes and save lots of time in post-production.

wassup josh get a chance to listen to that BE tape yet? little bassy eh? good ol cicero's FOH will do that i guess even with XY. need to get them hyper caps NOW!

Brian

[Simp-Dawg]

yeah bri, i gave it a little listen, even transferred it to my hd but still need to track it and all....yeah it was a bit bassy, i think maybe the wmod ua-5 contributed to that but i'm not sure, would like to hear a pmod in the same situation.  also, i have discovered i really do not care much for running the oktavas in xy, gives a pretty narrow stereo image IMHO.  i really like din and have yet to try ortf but i will give that a go one of these days.
i think i will have a chance to run the oktavas through an mp-2 this friday at jeff austin and friends in denver, i'm interested to hear it.

[Brian]

bill burge said he liked his oktava>mp2 recordings. as for stereo image, there pretty much is none at cicero's cause it's a mono PA and a narrow room.  and any other config just has waaay too much bass.  once i get some XLR>transformers>rcas i will be able to do matrixes again there. i think the hypers will help out a  lot in that room too.

Brian

[dklein]

hey - if you're with the band, cabling shouldn't be a problem.  Just use the snake.  I've set up a stage mic, run it through the snake and conveniently pick up my signal back at the board where I'm plugging in.  Beautiful!

[jctibbitts]

yeah that's what i was planning on doing...thanks all

now just need $500 for the oade w mod UA5 (im not screwing with doing the mod myself, i'll fck it up)

[joemango]

I did a matrix of Deep Banana Blackout with my UA-5(wmod) at Toad's place a couple weeks ago and it came out great, thanks to E (their sound man).  If the board sounded crappy, I would have had to live with it.  Lucky he knew what he was doing.

Another time I tried it with 2 separate sources and decks, and was never able to get the tracks to match up (I'm using SoundForge 6).  One track was about 2 seconds longer than the other due to differences in the A/D converter clocks.  I imagine that unless you have a digital mixer  that synchs this would always be a problem.

Does anyone know of a program that will reliably time-shift a wav file?

[blu666z]

Hey Alex.  I did a matrix in CEP a few weeks back.  

Lay both tracks in.  Played them together.  When I could start hearing an echo, I would stop the playback.  Split the SBD track, realign it(right side of the split) with the Aud track.  Mainly just looked for snare hits/peaks to use as a refrence.  Once I have it lined back up, lock the new region of the SBD track and start the process over againing listening for the 'echo.'  Really impressed with the results.  Granted I put about 8 hours into a 1hour becasue I went back and listened to the whole show several times before I was completely happy.  

-Kevin

[dklein]

I've probably done 15 or so over the last 6 months and have settled on a method.  My 2 devices drift too much for cutting between tracks to match sources (drift is ~10 ms/minute).

To my ear, once you're out by a few milliseconds, phasey things start happening.   It's easiest to hear in the vocals.  I tried a bunch of different stretch applications and found the one in Wavelab to sound excellent.  It took a little number crunching the first time around but I now know the stretch factor for the 2 devices and it's always the same (either 100.018 or 99.982 depending on which one I'm matching).

In wavelab, go process, timestretch, set to high quality and preserve pitch.  Then play with the ratio setting - you can use the other settings but I've discovered that the only accurate predictor of the file length comes from the ratio setting (up to 3 decimals).  Because of that, it's not worth trying to set it anywhere else.  

I have wavelab set to use 32 bit temp files - this may help the sound quality.

So you can either start by determining the gap between 2 points and calculating your desired stretch, or just zero in by trial and error.  If you go the latter way, setting the different stretches up in multitrack is a good way to look at them to guess where to go next.

When I do a matrix I usually have one 16 bit source (soundboard>jb3) and one 24 bit source (mics>laptop).  I've stretched each one to match the other.  I'm not sure which is best to leave and which should be stretched.  You could argue doing it tothe 24 bit source because it may provide more accurate data to work with.  At the same time, why touch the better source?  In any case, to my ears, the wavelab stretch sounds good.

Then I save it in 24 bit format and load the tracks in Cool Edit Pro multitrack.  Line 'em up, adjust amplitude, do any eq (often a bass rolloff) and mixdown to 32 bit.  Fades.  Save in 32 bit fp.  Back to wavelab for UV22HR dithering.

So there are many methods and software.  This is one stretch approach.  The other more common one is to edit between each track to realign the waves.  You delete during the gaps between songs where the edit will not be noticeable.  You want to get smooth edits (I know CEP has an auto smoothing feature that will blend smoothly with nearby samples).

Either way, it's guaranteed to be a time consuming process... but the results can be great if you get it right!

[joemango]

I'm gonna give it a go with SONAR 3.  I'll post results and techniques in a new thread, probably after the weekend.

[spearheadtaper]

its seems almost all sbd  > aud matrix that have been attemped ....has some serious echo from the sbd...whats the best way to remedy the ecoey sbd